> In theory GPT without sampling could still give good results if it used a search to look ahead, but it appears that even extraordinary amounts of computation

> In theory GPT without sampling could still give good results if it used a search to look ahead, but it appears that even extraordinary amounts of computation for look-ahead still is a long way from reaching the correct distribution, presumably because the exponential fan out is so fast that even 'huge' amounts of lookahead are still only testing a tiny fraction of the space. This is a reasonable guess, but unfortunately, it turns out to be more fundamental than just 'search is expensive'; there's something pathological in the model tree of completions that leads to degeneration, and definitely does not lead to the best possible completion, the more compute you use (nodes explore). If you use something like beam search, the more you search, the more likely you are to get trapped in one of the notorious repetition traps where it prints 'the the the'. This is a long-standing problem in NMT, and IIRC, I recall reading a paper where they went to the trouble of doing full Viterbi-style brute force to get the exact optimal result (according to the model) to avoid incomplete search possibly screwing things up, and the result were all still garbage. There are a couple theories why, but nothing I consider definitive. You can boost models a lot with limited search (best-of can make a big difference) and with other tricks which ought to be sorta equivalent (self-distillation and InstructGPT's RL finetuning), but it's never clear what tricks will work in advance. So, we'll see! I have a hunch it may just be, like adversarial examples, another blessing-of-scale in waiting.